0
$\begingroup$

I want to know more about the latest attempts to create synthesizers for singing. I would ideally like to know more about both the most widely respected algorithms/models as well as mainstream software or programs that implement them (whether free or proprietary). Can someone offer me resources that would give me a decent grasp of the above?

I chose to post this question on Signal Processing Stack Exchange since I know such software relies to some degree on signal processing type methods for analyzing or creating waves.

I'm very new to this area of research so any suggestions for where to start would be welcomed.

Thanks.

$\endgroup$

closed as too broad by lennon310, jojek, Deve, MBaz, Phonon Jan 5 '15 at 22:52

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

1
$\begingroup$

Well, there is sampling of singing voices. it's not the latest. one thing that you perhaps need to worry about is that, unless you get a sample root for each note going up the MIDI note scale, then you will likely be pitch shifting the notes up or down a little. it turns out that this causes a noticeable distortion sometimes called "munchkinization" or the "Alvin the Chipmunk" effect. to avoid that, each cycle of the quasi-periodic note needs to be isolated into little "wavelets" (not to be confused by the wavelet transform) or "grains" or "FOFs". then you would output the grains or FOFs at a rate of the fundamental frequency of the note without stretching or scrunching the grain (this requires overlapping and adding). there's an old paper by Keith Lent that discusses pitch shifting vocal notes without shifting formants, and another paper by me that analyzes the algorithm a little more rigorously. the latter, i can send a copy.

but for synthesis, you need not perform the real-time pitch detection that you would have to for pitch-shifting the voice live. what you can do is sample some notes from these sung voices and perform analysis on them off-line and generate a file of these grains or FOFs and store them in your synthesizer.

$\endgroup$
  • $\begingroup$ Very interesting. I know of the effect your talking but besides that all of that is new to me and very useful. That would be great! I'd love to see it. Email is got.lepton@gmail.com. Thanks! $\endgroup$ – Stan Shunpike Dec 7 '14 at 7:04

Not the answer you're looking for? Browse other questions tagged or ask your own question.